World J Surg DOI 10.1007/s00268-014-2617-9
Japanese D3 Lymph Node Dissection in Low Rectal Cancer with Inferior Mesenteric Lymph Node Metastases Mamiko Ubukata • Michio Itabashi Shingo Kameoka
•
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background There is currently no consensus on the significance of Japanese D3 lymph node dissection in low rectal cancer with inferior mesenteric lymph node (IMLN) metastasis. This is partly because, despite a number of studies on the subject, cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs. A retrospective study involving a large number of patients was conducted. Methods The subjects were 2,743 patients registered in the national registry of the Japanese Society for Cancer of the Colon and Rectum. The data were analyzed for (1) prognostic factors for IMLN metastasis, and (2) outcomes in R0 cases with IMLN metastasis. Results In the control group, 67 patients (2.7 %) were considered positive for IMLN metastasis. The outcomes in the 35 R0 cases with IMLN metastasis were 50.8 % for 5-year relapse-free survival (RFS) and 61.9 % for 5-year overall survival (OS), which were each better than for R1?R2 cases (5-year RFS 16.1 %, p = 0.0001; 5-year OS 26.7 %, p = 0.0002). The outcomes for R0 cases (total metastatic lymph nodes C7) with IMLN metastasis (5-year RFS 53.9 %, 5-year OS 68.8 %) did not differ significantly from those for IMLN(-) cases (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621). Conclusions It was confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large total number of metastatic lymph nodes, but if curative resection can be performed, a good prognosis can be expected. These
M. Ubukata (&) M. Itabashi S. Kameoka Department of Surgery 2, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan e-mail: [email protected]
results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis.
Introduction The current Japanese Classification of Colorectal Carcinoma [1] defines inferior mesenteric lymph nodes (IMLN) as ‘‘Lymph nodes along the inferior mesenteric artery proximal to the origin of the left colic artery’’, and states that these lymph nodes should be dissected when performing D3 dissection for rectal cancer and sigmoid colon cancer. However, there is currently no consensus on the effect of D3 dissection in low rectal cancer. This is because, despite a number of studies on the subject, cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs [2–12]. A retrospective study of the results of lymph node dissection in low rectal cancer with IMLN metastasis was conducted in a large number of patients using information from the Japanese Society for Cancer of the Colon and Rectum (JSCCR) registry.
Subjects and methods The subjects comprised 2,447 of the 2,743 patients included in the national registry of the JSCCR. This number excluded patients who had undergone trans-anal tumor excision and/or polypectomy. Furthermore, 1,128 of the 2,447 patients actually underwent D3 dissection. The following items were investigated: (1) prognostic factors in cases with IMLN metastasis; and (2) outcome in R0 cases (total metastatic lymph nodes C7) with IMLN metastasis.
123
World J Surg Table 1 Characteristics of low rectal cancer with inferior mesenteric lymph node metastases IMLN(?) (n = 67)
IMLN(-) (n = 2,380)
Sex Male Female Surgical procedure
1,585 (66.6)
21 (31.3)
795 (33.4) 0.2107:NS
APR
27 (40.3)
LAR
27 (40.3)
840 (35.3)
Har
0 (0)
86 (3.5):
936 (39.3)
Other
13 (19.4)
518 (21.8)
Macroscopic type
0.0108
0
0 (0)
219 (8.9)
1
3 (4.5)
168 (7.1)
2
50 (74.6)
1,573 (66.1)
3
12 (17.9)
193 (8.1)
4
0 (0)
14 (0.6)
5
0 (0)
44 (1.8)
Unknown
2 (3.0)
169 (7.1)
Tis T1
0 (0) 2 (3.0)
52 (21.3) 245 (10.3)
T2
11 (16.4)
562 (23.6)
T3
49 (73.1)
1,283 (54.3)
T4
5 (7.5)
128 (5.4)
Unknown
0 (0)
110 (4.6)
tub1
12 (14.9)
687 (28.9)
tub2
43 (64.2)
1,026 (43.1)
por
6 (9.0)
73 (3.1)
Depth of invasion
0.0658:NS
Histological type
0.0659:NS
muc
4 (6.0)
66 (2.8)
sig
1 (1.5)
6 (0.3)
Unknown
1 (1.5)
522 (21.9)
0 (0)
1,573 (64.3)
1
3 (4.5)
237 (10.0)
2–3
3 (4.5)
242 (10.2)
4–6
9 (13.4)
156 (6.6)
C7
52 (77.6)
172 (7.2)
5 (7.5)
91 (3.8)
Pre-operative radiotherapy
13 (19.4):
188 (7.9)
R2
18 (26.9)
219 (9.2)
Unknown
0 (0)
210 (8.8)
Data are presented as number of cases (%) APR abdominoperineal resection, Har Hartmann procedure, LAR low anterior resection, muc mucinous, NS not significant, por poorly differentiated, sig signet-ring cell, Tis carcinoma in situ, tub tubular
The significance of differences between pairs of groups was tested using the v2 test, and the Kaplan–Meier method was used for analysis of outcomes. Wilcoxon’s method was used for comparison of results; significance was set at p \ 0.05. Cases were recorded using the following abbreviations, in accordance with the Japanese Classification of Colorectal Carcinoma [1]: IMLN (inferior mesenteric lymph nodes) and low rectal cancer (cancer that is located in a segment of the large intestine between the peritoneal reflection and the superior border of the puborectal muscle). The location is confirmed by barium enema examination. We judge the height of the middle valve of Houston as the peritoneal reflection. In Japan, lateral lymph nodes are treated as regional lymph nodes. Complete dissection of all regional lymph nodes is recorded as D3 dissection. In rectal cancer, D3 represents lymph node dissection along the superior rectal and inferior mesenteric arteries and bilateral lateral lymph node dissection. Therefore, in the present study, we confirm R0 resection in patients with IMLN involvement who undergo D3 dissection (with no metastasis to para aortic lymph nodes and other organs by pre-operative/intra-operative diagnosis), even if they have lateral lymph node involvement.
Prognostic factors in cases with IMLN metastasis
0.1882:NS
None
61 (91.0)
2,190 (92.0)
Unknown
1 (1.5)
99 (4.2)
Done
18 (26.9)
700 (29.4)
None
48 (71.6)
1,581 (66.4)
Unknown
1 (1.5)
99 (4.2)
36 (53.7)
1,763 (74.1)
Post-operative chemotherapy
123
R1
p value
Results
0
Curability R0
IMLN(-) (n = 2,380)
\0.0001
Number of metastatic lymph nodes
Done
IMLN(?) (n = 67)
p value 0.7243:NS
46 (68.7)
Table 1 continued
0.4620:NS
\0.0001
In the control group, 67 patients (2.7 %) were judged to be positive for IMLN metastasis (time of relapse unknown: one patient). Compared with normal low rectal cancer, IMLN metastasis patients were characterized by more macroscopic type 3 tumors and a greater total number of metastatic lymph nodes, and they also tended to have lower levels of curability with curative resection (Table 1). Univariate analysis of prognostic factors in 66 patients with IMLN metastasis identified depth of invasion, preoperative carcinoembryonic antigen (CEA) value, and curability as poor prognostic factors for 5-year relapse-free survival (RFS), and surgical procedure, macroscopic type, depth of invasion, pre-operative CEA value, and curability
World J Surg Table 2 Univariate analysis of factors associated with 5-year relapse-free survival and 5-year overall survival (IMLN(?) 66 cases) N
5-year RFS
p value
5-year OS
Table 3 Multivariate analysis of factors associated with 5-year relapse-free survival and 5-year overall survival (IMLN(?) 66 cases)
p value
N
HR (95 % CI)
p value
5-year RFS Sex
0.9088:NS
0.9521:NS
Male
46
0.326
0.441
Female
20
0.389
0.495
Surgical procedure
0.1494:NS
0.0479
APR
27
0.185
0.264
LAR
26
0.423
0.572
Others
13
0.528
0.6
49
0.423
Macroscopic type Type 2 Other
0.0595:NS 17
0.118
Depth of invasion
0.0119 0.557 0.177
0.0014
0.0069
Tis/T1/T2
13
0.671
0.755
T3/T4
43
0.264
0.382
Histological type
0.9569:NS 56
0.353
0.485
por/muc/sig
10
0.3
0.3
Number of metastatic lymph nodes
0.6043:NS
1–6
15
0.305
C7
51
0.353
Pre-operative CEA value
0.381 0.0016
58
0.375
0.503
Cx8
8
0.125
0.125
Circumference of the bowel
0.4647:NS
1
43
2.48 (1.10–6.65)
\x8
58
Cx8 Curability
8
1 2.37 (1.00–4.99)
35
1
R1/R2
31
2.51 (1.38–4.64)
0.0026
5-year OS APR
27
1.77 (0.70–5.13)
0.2346:NS*
LAR
26
1.06 (0.41–3.07)
0.9016:NS**
Other
13
1
Type 2
49
1
Other
17
1.50 (0.70–3.09)
Tis/T1/T2
13
1
T3/T4
43
2.00 (0.80–6.06)
Pre-operative CEA value \x8 58 0.7858:NS
Cx8
8
0.2931:NS
0.1442:NS
1 2.04 (0.81–4.70)
0.1233:NS
Curability 26
0.414
0.453
R0
35
1
C1/2 circ
40
0.3
0.459
R1/R2
31
1.73 (0.83–3.67)
Location of the bowel
0.0511:NS
Anterior
13
0.615
Other
43
0.277
Pre-operative radiotherapy
0.0936:NS 0.692 0.397
0.3689:NS
0.4349:NS
0.1442:NS
APR abdominoperineal resection, CEA carcinoembryonic antigen, CI confidence interval, HR hazard ratio, IMLN inferior mesenteric lymph node, LAR low anterior resection, NS not significant, OS overall survival, RFS relapse-free survival, Tis carcinoma in situ *p value of HR between ‘‘APR’’ and ‘‘Other’’
Done
5
None
60 1
0.2
0.4
0.347
0.455
1
Post-operative chemotherapy
0.2224:NS
18
0.208
None
47
0.383
0.495
1
1
1
Curability
**p value of HR between ‘‘LAR’’ and ‘‘Other’’
1 0.4903:NS
Done Unknown
0.0504:NS
R0
\1/2 circ
Unknown
0.027
Pre-operative CEA value
Depth of invasion
0.479
\x8
13
T3/T4
Macroscopic type 0.9678:NS
0.0012
Tis/T1/T2
Surgical procedure 0.2494:NS
tub1/tub2
Depth of invasion
0.333
0.0001
0.0002
R0
35
0.508
0.619
R1/R2
31
0.161
0.267
APR abdominoperineal resection, CEA carcinoembryonic antigen, Har Hartmann procedure, IMLN inferior mesenteric lymph node, LAR low anterior resection, muc mucinous, NS not significant, OS overall survival, por poorly differentiated, RFS relapse-free survival, sig signet-ring cell, Tis carcinoma in situ, tub tubular
as poor prognostic factors for 5-year overall survival (OS) (Table 2). A multivariate analysis of these factors confirmed significant differences for depth of invasion and curability for 5-year RFS. Of these two factors, curability had a high hazard ratio (HR) of 2.51 and was judged to be an important prognostic factor (p = 0.0026) (Table 3). The outcomes in the 35 R0 cases with IMLN metastasis were 50.8 % for 5-year RFS and 61.9 % for 5-year OS, and both were better than the outcomes for R1?R2 cases (5year RFS 16.1 %, 5-year OS 26.7 %) (RFS: p = 0.0001, OS: p = 0.0002). These results confirm that a good outcome can be expected, even with IMLN metastasis, if curative resection can be performed (Fig. 1).
123
World J Surg
IMLN(-) group (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621).
Discussion
Fig. 1 Outcomes (RFS) of low rectal cancer (R0) with inferior mesenteric lymph node metastases. The outcomes in the 35 R0 cases with inferior mesenteric lymph node metastasis are better (5-year RFS 50.8 %) than for R1?R2 cases (16.1 %), (RFS: p = 0.0001). IMLN inferior mesenteric lymph node, RFS relapse-free survival
Outcomes in R0 cases (total metastatic lymph nodes C 7) with IMLN metastasis The total number of metastatic lymph nodes in the 35 R0 patients with IMLN metastasis tended to be high, with one to three metastatic lymph nodes in four patients, four to six in five patients, and seven or more in 26 patients (Fig. 2). The prognosis was compared between IMLN(?) and IMLN(-) patients, including only those with a total of seven or more metastatic lymph nodes. The results regarding prognosis showed that the 5-year RFS was 53.9 % and the 5-year OS was 68.8 % in the IMLN(?) group, which was not significantly different from the
Fig. 2 Outcomes (RFS) of low rectal cancer (R0) with inferior mesenteric lymph node metastases (seven or more metastatic lymph nodes). The prognosis was compared between IMLN(?) and IMLN(-) patients, including only those with seven or more total metastatic lymph nodes. The results show that the 5-year RFS is 53.9 %, which is not significantly different from the IMLN(-) group (54.6 %) (p = 0.9515). IMLN inferior mesenteric lymph node, RFS relapse-free survival
123
Because cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs, there has been insufficient prognostic analysis of such cases and no consensus on the value of R0 resection [2–12]. Until now, very few studies have been published on a large number of patients undergoing D3 lymph node dissection for low rectal cancer. We therefore considered that new studies, such as the present study, using national data on numerous cases from multiple institutions are necessary. In the present study, 67 patients who were positive for IMLN metastasis were analyzed, using information from the JSCCR registry, to clarify the effect of lymph node dissection in low rectal cancer with IMLN metastasis. According to several studies, 0.3–8.6 % of patients with rectal cancer are positive for IMLN metastasis [2–9], and the figure from the JSCCR registry was also low, at 2.7 %. In a number of outcome studies, 5-year OS was 14–40 %, and 5-year RFS was 13.8–31 % in all rectal cancer patients with IMLN metastasis [2–6]. Uehara et al. [2] reported very poor outcomes in cases of IMLN metastasis limited to low rectal cancer, with a 5-year OS of 0 %, but the study only analyzed four patients with IMLN metastasis. Thus, studies with a greater number of cases were required. In previous studies of the value of IMLN dissection, evidence suggested that the outcome was poor in cases of IMLN metastasis, and that dissection had little effect on outcomes due to the low rate of metastasis [2–4, 8], while other studies claimed that IMLN metastasis is not a factor for poor prognosis and that dissection is important [5–7, 9, 11]. One of these studies suggested that IMLN is closer to the M category than to the regional lymph nodes [4]. The current study found that, in the 35 R0 patients with low rectal cancer and IMLN metastasis, the 5-year OS was 61.9 % and the 5-year RFS was 50.8 % (for all 66 cases with IMLN metastasis, the 5-year OS was 46.5 % and the 5-year RFS was 34.5 %), which confirmed that a good outcome can be expected, even with IMLN metastasis, if curative resection can be performed. We concluded that IMLN in low rectal cancer can be handled as regional lymph nodes that need to be dissected according to the extent of metastasis. After studying the form of recurrence, we were able to confirm that there were no more cases of lymph node recurrence in IMLN(?) R0 cases than in IMLN(-) R0 cases. Lange et al. [3] reported that outcomes were significantly poorer in cases of rectal cancer with five or more metastatic lymph nodes. Therefore, the present study compared the
World J Surg
prognosis between IMLN(?) and IMLN(-) patients, including only those with a total of seven or more metastatic lymph nodes. The results showed that the 5-year RFS was 53.9 % and the 5-year OS was 68.8 % in the IMLN(?) group, which was not significantly different from the IMLN(-) group (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621). Therefore, we concluded that, even in patients with seven or more metastatic lymph nodes, a good outcome could be expected if curative resection were possible. Additionally, in Western countries, pre-operative chemotherapy/radiation therapy is commonly performed in patients with low rectal cancer. However, this type of therapy is yet to be used as the standard of care in Japan. Nonetheless, in the present study, we investigated the effects on prognosis of pre-operative radiation therapy/postoperative chemotherapy. On univariate analysis, the prognosis was compared between those who underwent preoperative radiation therapy and those who did not, but there were no significant differences in prognosis between the two groups (RFS: p = 0.3689, OS: p = 0.4349) (Table 2). The prognosis was also compared between patients who underwent post-operative chemotherapy and those who did not on univariate analysis, and no significant differences were found (RFS: p = 0.4903, OS: p = 0.2224) (Table 2). We believe that, indeed, R0 resection surgery is extremely crucial as a prognostic factor in low rectal cancer. In this study, outcomes were compared in R0 cases with IMLN metastasis with and without lateral lymph node metastasis, but there was no clear difference in outcomes between groups; the 5-year RFS was 41.7 % in lateral lymph node metastasis (-) cases and 55.6 % in lateral lymph node metastasis(?) cases (p = 0.6994). In Western countries, low rectal cancer is usually defined as ‘‘rectal cancer that is located within 6–7 cm from the anal verge’’ [2, 13]. Conversely, in the present study, we used a definition in accordance with the Japanese Classification of Colorectal Carcinoma [1]. In Japan, the definition of low rectal cancer is ‘‘cancer that is located in a segment of the large intestine between the peritoneal reflection and the superior border of the puborectal muscle.’’ Therefore, this difference should be considered when evaluating this study. This was a retrospective study and may have included biases in background factors, because allocation to the IMLN dissection group and non-dissection group was not randomized. A controlled trial with patients randomized to dissection and non-dissection would be difficult to conduct in Japan at present. It is therefore necessary to accumulate more cases and study the data, taking background factors into account. The present study confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large number
of total metastatic lymph nodes, but a good outcome can be expected with curative resection. Thus, these results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis. Conclusions We confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large total number of metastatic lymph nodes, but a good outcome can be expected if curative resection can be performed. IMLN in low rectal cancer can be treated as regional lymph nodes from the perspective of disease progression and treatment. These results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis. Acknowledgments The authors would like to thank the Japanese Society for Cancer of the Colon and Rectum, as well as the other staff at JSCCR. No grant support was received for this article. The authors have no conflicts of interest that are relevant to the content of this article. Information for this study was provided by the Japanese Society for Cancer of the Colon and Rectum. Conflict of interest
None
References 1. Japanese Society for Cancer of the Colon and Rectum (2009) Japanese classification of colorectal carcinoma, 2nd edn. Kanehara & Co., Ltd, Tokyo 2. Uehara K, Yamamoto S, Fujita S et al (2007) Impact of upward lymph node dissection on survival rates in advanced lower rectal carcinoma. Dig Surg 24:375–381. doi:10.1159/000107779 3. Lange MM, Buunen M, van de Velde CJ et al (2008) Level of arterial ligation in rectal cancer surgery: low tie preferred over high tie. A review. Dis Colon Rectum 51:1139–1145. doi:10. 1007/s10350-008-9328-y 4. Kim JC, Lee KH, Yu CS et al (2004) The clinicopathological significance of inferior mesenteric lymph node metastasis in colorectal cancer. Eur J Surg Oncol 30:271–279. doi:10.1245/ s10434-010-1291-x 5. Chin CC, Yeh CY, Tang R et al (2008) The oncologic benefit of high ligation of the inferior mesenteric artery in the surgical treatment of rectal or sigmoid colon cancer. Int J Colorectal Dis 23:783–788. doi:10.1007/s00384-008-0465-5 6. Kanemitsu Y, Hirai T, Komori K et al (2006) Survival benefit of high ligation of the inferior mesenteric artery in sigmoid colon or rectal cancer surgery. Br J Surg 93:609–615. doi:10.1245/s10434010-1291-x 7. Yi JW, Lee TG, Lee HS et al (2012) Apical-node metastasis in sigmoid colon or rectal cancer: is it a factor that indicates a poor prognosis after high ligation? Int J Colorectal Dis 27:81–87. doi:10.1007/s00384-011-1271-z 8. Kim JS, Sohn DK, Park JS et al (2011) Prognostic significance of distribution of lymph node metastasis in advanced mid or low
123
World J Surg rectal cancer. J Surgical Oncol 104:486–492. doi:10.1002/jso. 21966 9. Alici A, Kement M, Gezen C et al (2010) Apical lymph nodes at the root of the inferior mesenteric artery in distal colorectal cancer: an analysis of the risk of tumor involvement and the impact of high ligation on anastomotic integrity. Tech Coloproctol 14:1–8. doi:10.1007/s10151-009-0547-6 10. Kawamura YJ, Umetani N, Sunami E et al (2000) Effect of high ligation on the long-term result of patients with operable colon cancer, particularly those with limited nodal involvement. Eur J Surg 166:803–807. doi:10.1080/110241500447443
123
11. Titu LV, Tweedle E, Rooney PS (2008) High tie of the inferior mesenteric artery in curative surgery for left colonic and rectal cancers: a systematic review. Dig Surg 25:148–157. doi:10.1159/000128172 12. Hida JI, Okuno K (2013) High ligation of the inferior mesenteric artery in rectal cancer surgery. Surg Today 43:8–19. doi:10.1007/ s00595-012-0359-6 13. Eric R, Quentin D, Veronique V (2013) Low rectal cancer: classification and standardization of surgery. Dis Colon Rectum 56:560–567. doi:10.1097/DCR.0b013e31827c4a8c
Japanese D3 Lymph Node Dissection in Low Rectal Cancer with Inferior Mesenteric Lymph Node Metastases Mamiko Ubukata • Michio Itabashi Shingo Kameoka
•
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background There is currently no consensus on the significance of Japanese D3 lymph node dissection in low rectal cancer with inferior mesenteric lymph node (IMLN) metastasis. This is partly because, despite a number of studies on the subject, cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs. A retrospective study involving a large number of patients was conducted. Methods The subjects were 2,743 patients registered in the national registry of the Japanese Society for Cancer of the Colon and Rectum. The data were analyzed for (1) prognostic factors for IMLN metastasis, and (2) outcomes in R0 cases with IMLN metastasis. Results In the control group, 67 patients (2.7 %) were considered positive for IMLN metastasis. The outcomes in the 35 R0 cases with IMLN metastasis were 50.8 % for 5-year relapse-free survival (RFS) and 61.9 % for 5-year overall survival (OS), which were each better than for R1?R2 cases (5-year RFS 16.1 %, p = 0.0001; 5-year OS 26.7 %, p = 0.0002). The outcomes for R0 cases (total metastatic lymph nodes C7) with IMLN metastasis (5-year RFS 53.9 %, 5-year OS 68.8 %) did not differ significantly from those for IMLN(-) cases (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621). Conclusions It was confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large total number of metastatic lymph nodes, but if curative resection can be performed, a good prognosis can be expected. These
M. Ubukata (&) M. Itabashi S. Kameoka Department of Surgery 2, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan e-mail: [email protected]
results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis.
Introduction The current Japanese Classification of Colorectal Carcinoma [1] defines inferior mesenteric lymph nodes (IMLN) as ‘‘Lymph nodes along the inferior mesenteric artery proximal to the origin of the left colic artery’’, and states that these lymph nodes should be dissected when performing D3 dissection for rectal cancer and sigmoid colon cancer. However, there is currently no consensus on the effect of D3 dissection in low rectal cancer. This is because, despite a number of studies on the subject, cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs [2–12]. A retrospective study of the results of lymph node dissection in low rectal cancer with IMLN metastasis was conducted in a large number of patients using information from the Japanese Society for Cancer of the Colon and Rectum (JSCCR) registry.
Subjects and methods The subjects comprised 2,447 of the 2,743 patients included in the national registry of the JSCCR. This number excluded patients who had undergone trans-anal tumor excision and/or polypectomy. Furthermore, 1,128 of the 2,447 patients actually underwent D3 dissection. The following items were investigated: (1) prognostic factors in cases with IMLN metastasis; and (2) outcome in R0 cases (total metastatic lymph nodes C7) with IMLN metastasis.
123
World J Surg Table 1 Characteristics of low rectal cancer with inferior mesenteric lymph node metastases IMLN(?) (n = 67)
IMLN(-) (n = 2,380)
Sex Male Female Surgical procedure
1,585 (66.6)
21 (31.3)
795 (33.4) 0.2107:NS
APR
27 (40.3)
LAR
27 (40.3)
840 (35.3)
Har
0 (0)
86 (3.5):
936 (39.3)
Other
13 (19.4)
518 (21.8)
Macroscopic type
0.0108
0
0 (0)
219 (8.9)
1
3 (4.5)
168 (7.1)
2
50 (74.6)
1,573 (66.1)
3
12 (17.9)
193 (8.1)
4
0 (0)
14 (0.6)
5
0 (0)
44 (1.8)
Unknown
2 (3.0)
169 (7.1)
Tis T1
0 (0) 2 (3.0)
52 (21.3) 245 (10.3)
T2
11 (16.4)
562 (23.6)
T3
49 (73.1)
1,283 (54.3)
T4
5 (7.5)
128 (5.4)
Unknown
0 (0)
110 (4.6)
tub1
12 (14.9)
687 (28.9)
tub2
43 (64.2)
1,026 (43.1)
por
6 (9.0)
73 (3.1)
Depth of invasion
0.0658:NS
Histological type
0.0659:NS
muc
4 (6.0)
66 (2.8)
sig
1 (1.5)
6 (0.3)
Unknown
1 (1.5)
522 (21.9)
0 (0)
1,573 (64.3)
1
3 (4.5)
237 (10.0)
2–3
3 (4.5)
242 (10.2)
4–6
9 (13.4)
156 (6.6)
C7
52 (77.6)
172 (7.2)
5 (7.5)
91 (3.8)
Pre-operative radiotherapy
13 (19.4):
188 (7.9)
R2
18 (26.9)
219 (9.2)
Unknown
0 (0)
210 (8.8)
Data are presented as number of cases (%) APR abdominoperineal resection, Har Hartmann procedure, LAR low anterior resection, muc mucinous, NS not significant, por poorly differentiated, sig signet-ring cell, Tis carcinoma in situ, tub tubular
The significance of differences between pairs of groups was tested using the v2 test, and the Kaplan–Meier method was used for analysis of outcomes. Wilcoxon’s method was used for comparison of results; significance was set at p \ 0.05. Cases were recorded using the following abbreviations, in accordance with the Japanese Classification of Colorectal Carcinoma [1]: IMLN (inferior mesenteric lymph nodes) and low rectal cancer (cancer that is located in a segment of the large intestine between the peritoneal reflection and the superior border of the puborectal muscle). The location is confirmed by barium enema examination. We judge the height of the middle valve of Houston as the peritoneal reflection. In Japan, lateral lymph nodes are treated as regional lymph nodes. Complete dissection of all regional lymph nodes is recorded as D3 dissection. In rectal cancer, D3 represents lymph node dissection along the superior rectal and inferior mesenteric arteries and bilateral lateral lymph node dissection. Therefore, in the present study, we confirm R0 resection in patients with IMLN involvement who undergo D3 dissection (with no metastasis to para aortic lymph nodes and other organs by pre-operative/intra-operative diagnosis), even if they have lateral lymph node involvement.
Prognostic factors in cases with IMLN metastasis
0.1882:NS
None
61 (91.0)
2,190 (92.0)
Unknown
1 (1.5)
99 (4.2)
Done
18 (26.9)
700 (29.4)
None
48 (71.6)
1,581 (66.4)
Unknown
1 (1.5)
99 (4.2)
36 (53.7)
1,763 (74.1)
Post-operative chemotherapy
123
R1
p value
Results
0
Curability R0
IMLN(-) (n = 2,380)
\0.0001
Number of metastatic lymph nodes
Done
IMLN(?) (n = 67)
p value 0.7243:NS
46 (68.7)
Table 1 continued
0.4620:NS
\0.0001
In the control group, 67 patients (2.7 %) were judged to be positive for IMLN metastasis (time of relapse unknown: one patient). Compared with normal low rectal cancer, IMLN metastasis patients were characterized by more macroscopic type 3 tumors and a greater total number of metastatic lymph nodes, and they also tended to have lower levels of curability with curative resection (Table 1). Univariate analysis of prognostic factors in 66 patients with IMLN metastasis identified depth of invasion, preoperative carcinoembryonic antigen (CEA) value, and curability as poor prognostic factors for 5-year relapse-free survival (RFS), and surgical procedure, macroscopic type, depth of invasion, pre-operative CEA value, and curability
World J Surg Table 2 Univariate analysis of factors associated with 5-year relapse-free survival and 5-year overall survival (IMLN(?) 66 cases) N
5-year RFS
p value
5-year OS
Table 3 Multivariate analysis of factors associated with 5-year relapse-free survival and 5-year overall survival (IMLN(?) 66 cases)
p value
N
HR (95 % CI)
p value
5-year RFS Sex
0.9088:NS
0.9521:NS
Male
46
0.326
0.441
Female
20
0.389
0.495
Surgical procedure
0.1494:NS
0.0479
APR
27
0.185
0.264
LAR
26
0.423
0.572
Others
13
0.528
0.6
49
0.423
Macroscopic type Type 2 Other
0.0595:NS 17
0.118
Depth of invasion
0.0119 0.557 0.177
0.0014
0.0069
Tis/T1/T2
13
0.671
0.755
T3/T4
43
0.264
0.382
Histological type
0.9569:NS 56
0.353
0.485
por/muc/sig
10
0.3
0.3
Number of metastatic lymph nodes
0.6043:NS
1–6
15
0.305
C7
51
0.353
Pre-operative CEA value
0.381 0.0016
58
0.375
0.503
Cx8
8
0.125
0.125
Circumference of the bowel
0.4647:NS
1
43
2.48 (1.10–6.65)
\x8
58
Cx8 Curability
8
1 2.37 (1.00–4.99)
35
1
R1/R2
31
2.51 (1.38–4.64)
0.0026
5-year OS APR
27
1.77 (0.70–5.13)
0.2346:NS*
LAR
26
1.06 (0.41–3.07)
0.9016:NS**
Other
13
1
Type 2
49
1
Other
17
1.50 (0.70–3.09)
Tis/T1/T2
13
1
T3/T4
43
2.00 (0.80–6.06)
Pre-operative CEA value \x8 58 0.7858:NS
Cx8
8
0.2931:NS
0.1442:NS
1 2.04 (0.81–4.70)
0.1233:NS
Curability 26
0.414
0.453
R0
35
1
C1/2 circ
40
0.3
0.459
R1/R2
31
1.73 (0.83–3.67)
Location of the bowel
0.0511:NS
Anterior
13
0.615
Other
43
0.277
Pre-operative radiotherapy
0.0936:NS 0.692 0.397
0.3689:NS
0.4349:NS
0.1442:NS
APR abdominoperineal resection, CEA carcinoembryonic antigen, CI confidence interval, HR hazard ratio, IMLN inferior mesenteric lymph node, LAR low anterior resection, NS not significant, OS overall survival, RFS relapse-free survival, Tis carcinoma in situ *p value of HR between ‘‘APR’’ and ‘‘Other’’
Done
5
None
60 1
0.2
0.4
0.347
0.455
1
Post-operative chemotherapy
0.2224:NS
18
0.208
None
47
0.383
0.495
1
1
1
Curability
**p value of HR between ‘‘LAR’’ and ‘‘Other’’
1 0.4903:NS
Done Unknown
0.0504:NS
R0
\1/2 circ
Unknown
0.027
Pre-operative CEA value
Depth of invasion
0.479
\x8
13
T3/T4
Macroscopic type 0.9678:NS
0.0012
Tis/T1/T2
Surgical procedure 0.2494:NS
tub1/tub2
Depth of invasion
0.333
0.0001
0.0002
R0
35
0.508
0.619
R1/R2
31
0.161
0.267
APR abdominoperineal resection, CEA carcinoembryonic antigen, Har Hartmann procedure, IMLN inferior mesenteric lymph node, LAR low anterior resection, muc mucinous, NS not significant, OS overall survival, por poorly differentiated, RFS relapse-free survival, sig signet-ring cell, Tis carcinoma in situ, tub tubular
as poor prognostic factors for 5-year overall survival (OS) (Table 2). A multivariate analysis of these factors confirmed significant differences for depth of invasion and curability for 5-year RFS. Of these two factors, curability had a high hazard ratio (HR) of 2.51 and was judged to be an important prognostic factor (p = 0.0026) (Table 3). The outcomes in the 35 R0 cases with IMLN metastasis were 50.8 % for 5-year RFS and 61.9 % for 5-year OS, and both were better than the outcomes for R1?R2 cases (5year RFS 16.1 %, 5-year OS 26.7 %) (RFS: p = 0.0001, OS: p = 0.0002). These results confirm that a good outcome can be expected, even with IMLN metastasis, if curative resection can be performed (Fig. 1).
123
World J Surg
IMLN(-) group (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621).
Discussion
Fig. 1 Outcomes (RFS) of low rectal cancer (R0) with inferior mesenteric lymph node metastases. The outcomes in the 35 R0 cases with inferior mesenteric lymph node metastasis are better (5-year RFS 50.8 %) than for R1?R2 cases (16.1 %), (RFS: p = 0.0001). IMLN inferior mesenteric lymph node, RFS relapse-free survival
Outcomes in R0 cases (total metastatic lymph nodes C 7) with IMLN metastasis The total number of metastatic lymph nodes in the 35 R0 patients with IMLN metastasis tended to be high, with one to three metastatic lymph nodes in four patients, four to six in five patients, and seven or more in 26 patients (Fig. 2). The prognosis was compared between IMLN(?) and IMLN(-) patients, including only those with a total of seven or more metastatic lymph nodes. The results regarding prognosis showed that the 5-year RFS was 53.9 % and the 5-year OS was 68.8 % in the IMLN(?) group, which was not significantly different from the
Fig. 2 Outcomes (RFS) of low rectal cancer (R0) with inferior mesenteric lymph node metastases (seven or more metastatic lymph nodes). The prognosis was compared between IMLN(?) and IMLN(-) patients, including only those with seven or more total metastatic lymph nodes. The results show that the 5-year RFS is 53.9 %, which is not significantly different from the IMLN(-) group (54.6 %) (p = 0.9515). IMLN inferior mesenteric lymph node, RFS relapse-free survival
123
Because cases of IMLN metastasis are relatively rare, and there are few cases of curative resection because of metastasis to other organs, there has been insufficient prognostic analysis of such cases and no consensus on the value of R0 resection [2–12]. Until now, very few studies have been published on a large number of patients undergoing D3 lymph node dissection for low rectal cancer. We therefore considered that new studies, such as the present study, using national data on numerous cases from multiple institutions are necessary. In the present study, 67 patients who were positive for IMLN metastasis were analyzed, using information from the JSCCR registry, to clarify the effect of lymph node dissection in low rectal cancer with IMLN metastasis. According to several studies, 0.3–8.6 % of patients with rectal cancer are positive for IMLN metastasis [2–9], and the figure from the JSCCR registry was also low, at 2.7 %. In a number of outcome studies, 5-year OS was 14–40 %, and 5-year RFS was 13.8–31 % in all rectal cancer patients with IMLN metastasis [2–6]. Uehara et al. [2] reported very poor outcomes in cases of IMLN metastasis limited to low rectal cancer, with a 5-year OS of 0 %, but the study only analyzed four patients with IMLN metastasis. Thus, studies with a greater number of cases were required. In previous studies of the value of IMLN dissection, evidence suggested that the outcome was poor in cases of IMLN metastasis, and that dissection had little effect on outcomes due to the low rate of metastasis [2–4, 8], while other studies claimed that IMLN metastasis is not a factor for poor prognosis and that dissection is important [5–7, 9, 11]. One of these studies suggested that IMLN is closer to the M category than to the regional lymph nodes [4]. The current study found that, in the 35 R0 patients with low rectal cancer and IMLN metastasis, the 5-year OS was 61.9 % and the 5-year RFS was 50.8 % (for all 66 cases with IMLN metastasis, the 5-year OS was 46.5 % and the 5-year RFS was 34.5 %), which confirmed that a good outcome can be expected, even with IMLN metastasis, if curative resection can be performed. We concluded that IMLN in low rectal cancer can be handled as regional lymph nodes that need to be dissected according to the extent of metastasis. After studying the form of recurrence, we were able to confirm that there were no more cases of lymph node recurrence in IMLN(?) R0 cases than in IMLN(-) R0 cases. Lange et al. [3] reported that outcomes were significantly poorer in cases of rectal cancer with five or more metastatic lymph nodes. Therefore, the present study compared the
World J Surg
prognosis between IMLN(?) and IMLN(-) patients, including only those with a total of seven or more metastatic lymph nodes. The results showed that the 5-year RFS was 53.9 % and the 5-year OS was 68.8 % in the IMLN(?) group, which was not significantly different from the IMLN(-) group (5-year RFS 54.6 %, 5-year OS 57.1 %) (RFS: p = 0.9515, OS: p = 0.4621). Therefore, we concluded that, even in patients with seven or more metastatic lymph nodes, a good outcome could be expected if curative resection were possible. Additionally, in Western countries, pre-operative chemotherapy/radiation therapy is commonly performed in patients with low rectal cancer. However, this type of therapy is yet to be used as the standard of care in Japan. Nonetheless, in the present study, we investigated the effects on prognosis of pre-operative radiation therapy/postoperative chemotherapy. On univariate analysis, the prognosis was compared between those who underwent preoperative radiation therapy and those who did not, but there were no significant differences in prognosis between the two groups (RFS: p = 0.3689, OS: p = 0.4349) (Table 2). The prognosis was also compared between patients who underwent post-operative chemotherapy and those who did not on univariate analysis, and no significant differences were found (RFS: p = 0.4903, OS: p = 0.2224) (Table 2). We believe that, indeed, R0 resection surgery is extremely crucial as a prognostic factor in low rectal cancer. In this study, outcomes were compared in R0 cases with IMLN metastasis with and without lateral lymph node metastasis, but there was no clear difference in outcomes between groups; the 5-year RFS was 41.7 % in lateral lymph node metastasis (-) cases and 55.6 % in lateral lymph node metastasis(?) cases (p = 0.6994). In Western countries, low rectal cancer is usually defined as ‘‘rectal cancer that is located within 6–7 cm from the anal verge’’ [2, 13]. Conversely, in the present study, we used a definition in accordance with the Japanese Classification of Colorectal Carcinoma [1]. In Japan, the definition of low rectal cancer is ‘‘cancer that is located in a segment of the large intestine between the peritoneal reflection and the superior border of the puborectal muscle.’’ Therefore, this difference should be considered when evaluating this study. This was a retrospective study and may have included biases in background factors, because allocation to the IMLN dissection group and non-dissection group was not randomized. A controlled trial with patients randomized to dissection and non-dissection would be difficult to conduct in Japan at present. It is therefore necessary to accumulate more cases and study the data, taking background factors into account. The present study confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large number
of total metastatic lymph nodes, but a good outcome can be expected with curative resection. Thus, these results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis. Conclusions We confirmed that cases of IMLN metastasis in low rectal cancer tend to have a large total number of metastatic lymph nodes, but a good outcome can be expected if curative resection can be performed. IMLN in low rectal cancer can be treated as regional lymph nodes from the perspective of disease progression and treatment. These results demonstrate the value of radical Japanese D3 lymph node dissection in low rectal cancer with IMLN metastasis. Acknowledgments The authors would like to thank the Japanese Society for Cancer of the Colon and Rectum, as well as the other staff at JSCCR. No grant support was received for this article. The authors have no conflicts of interest that are relevant to the content of this article. Information for this study was provided by the Japanese Society for Cancer of the Colon and Rectum. Conflict of interest
None
References 1. Japanese Society for Cancer of the Colon and Rectum (2009) Japanese classification of colorectal carcinoma, 2nd edn. Kanehara & Co., Ltd, Tokyo 2. Uehara K, Yamamoto S, Fujita S et al (2007) Impact of upward lymph node dissection on survival rates in advanced lower rectal carcinoma. Dig Surg 24:375–381. doi:10.1159/000107779 3. Lange MM, Buunen M, van de Velde CJ et al (2008) Level of arterial ligation in rectal cancer surgery: low tie preferred over high tie. A review. Dis Colon Rectum 51:1139–1145. doi:10. 1007/s10350-008-9328-y 4. Kim JC, Lee KH, Yu CS et al (2004) The clinicopathological significance of inferior mesenteric lymph node metastasis in colorectal cancer. Eur J Surg Oncol 30:271–279. doi:10.1245/ s10434-010-1291-x 5. Chin CC, Yeh CY, Tang R et al (2008) The oncologic benefit of high ligation of the inferior mesenteric artery in the surgical treatment of rectal or sigmoid colon cancer. Int J Colorectal Dis 23:783–788. doi:10.1007/s00384-008-0465-5 6. Kanemitsu Y, Hirai T, Komori K et al (2006) Survival benefit of high ligation of the inferior mesenteric artery in sigmoid colon or rectal cancer surgery. Br J Surg 93:609–615. doi:10.1245/s10434010-1291-x 7. Yi JW, Lee TG, Lee HS et al (2012) Apical-node metastasis in sigmoid colon or rectal cancer: is it a factor that indicates a poor prognosis after high ligation? Int J Colorectal Dis 27:81–87. doi:10.1007/s00384-011-1271-z 8. Kim JS, Sohn DK, Park JS et al (2011) Prognostic significance of distribution of lymph node metastasis in advanced mid or low
123
World J Surg rectal cancer. J Surgical Oncol 104:486–492. doi:10.1002/jso. 21966 9. Alici A, Kement M, Gezen C et al (2010) Apical lymph nodes at the root of the inferior mesenteric artery in distal colorectal cancer: an analysis of the risk of tumor involvement and the impact of high ligation on anastomotic integrity. Tech Coloproctol 14:1–8. doi:10.1007/s10151-009-0547-6 10. Kawamura YJ, Umetani N, Sunami E et al (2000) Effect of high ligation on the long-term result of patients with operable colon cancer, particularly those with limited nodal involvement. Eur J Surg 166:803–807. doi:10.1080/110241500447443
123
11. Titu LV, Tweedle E, Rooney PS (2008) High tie of the inferior mesenteric artery in curative surgery for left colonic and rectal cancers: a systematic review. Dig Surg 25:148–157. doi:10.1159/000128172 12. Hida JI, Okuno K (2013) High ligation of the inferior mesenteric artery in rectal cancer surgery. Surg Today 43:8–19. doi:10.1007/ s00595-012-0359-6 13. Eric R, Quentin D, Veronique V (2013) Low rectal cancer: classification and standardization of surgery. Dis Colon Rectum 56:560–567. doi:10.1097/DCR.0b013e31827c4a8c
